I the thermal expansion problem is solved in Dubai, it also means that it can be solved on many more places. From a prototypal standpoint, it's a pretty good place.
It is far from clear that the thermal expansion problem can be solved at all. AFAICT no one has ever proposed a solution that is even remotely close to being viable.
Could you please show why very-low-speed, low-precision linear bearing to "allow longitudinal slip" of the tube past the pillars, is not "even remotely close to being viable". It's an honest question. I know very little about bearings, but I have yet to find a serious discussion about it anywhere.
Why do you keep posting this article? It can't be a critique of the free "longitudinal slip" linear bearings proposal, because the author assumed the tube is rigidly attached to the pillars. (And it is not.)